Corpora: error tagging of learners' English

John Milton lcjohn at ust.hk
Thu Nov 16 15:29:38 UTC 2000


Vicky,

No program will do what you want. If a program could automatically
and reliably parse interlanguage, it would hammer out the most glaring
infelicities, and there'd be no need for error analysis. Even the latest
version of MSWord is pretty hopeless at spotting the types of
ungrammaticalities common among HK learners.

Manfred Pienemann worked with a crew in the early 90s in Australia to
develop COALA, a tool for 'semi-automatic' interlanguage analysis (which
essentially meant incrementing annotations in a relational database).
However, I know of no major error analysis project that has used the
tool. They have a web page and some articles at
http://ausarts.anu.edu.au/mel/linguistics/welcome.html

John Milton
______________________________
John Milton
email: lcjohn at ust.hk
WordPilot: http://home.ust.hk/~autolang/
On Thu, 16 Nov 2000, vicky man wrote:

> Dear all,
>
> I'd like to know if there are any 'error tagging' softwares that can
> effectively indicate the kinds of errors found in learners' English?
>
> Your suggestions would be very much appreciated.
>
> all the best, Vicky
>
> *****
> Vicky Man
> Language Center
> Hong Kong Baptist University
> Kowloon Tong, Hong Kong
>
>
>



More information about the Corpora mailing list